With fiber-optic seismic acquisition development, continuous dense seismic monitoring is becoming increasingly more accessible. Repurposing fiber cables in telecommunication conduits makes it possible to run seismic studies at low cost, even in locations where traditional seismometers are not easily installed, such as in urban areas. However, due to the large volume of continuous streaming data, data collected in such a manner will go to waste unless we significantly automate the processing workflow. We train a convolutional neural network (CNN) for earthquake detection using data acquired over three years by fiber cables in telecommunication conduits under Stanford University campus. We demonstrate that fiber-optic systems can effectively complement sparse seismometer networks to detect local earthquakes. The CNN allows for reliable earthquake detection despite a low signal-to-noise ratio and even detects small-amplitude previously-uncataloged events. [link to publication]